Yesterday I attended the DPC event on preservation metadata which focused on the PREMIS and METS standards. Coming at a time when I am reviewing various systems and software for managing digital archives and research data this was quite timely. The main message I took away from the day is that the key to actually creating and managing this metadata as ever is having the necessary tools!
|Jelly Beans by kayaker1204, on Flickr|
- lots of these were consumed yesterday!
Earlier this year with the help of an intern we collected together digital media deposited with the Borthwick Institute and got the data safely stored on our digital archive file store. For the time being, the metadata I am holding about these files is pretty limited (I am currently working to level one of the NDSA Levels of Digital Preservation – metadata comes further down the line so I’m temporarily off the hook!).
I do have some technical metadata for files (MD5 checksums and output from DROID) however, the need to have a system in place to hold information about preservation actions is becoming more pressing. Working on the premis that doing something is better than doing nothing, there have been certain files that I have already been compelled to migrate to file formats more suitable for long term archiving. Documenting these kinds of actions is an essential part of any digital archivists job but my problem at the moment is that I have no proper system in which to hold this information.
The benefits of using PREMIS are clear. As a couple of the speakers at yesterday's event stated, it provides a metadata schema that includes all things “that most working preservation repositories are likely to need to know to support digital preservation”. It is flexible enough that you can pick and choose from the elements, using only those that you require. If this is the case then I do not see a reason to re-invent the wheel and design a new schema to record the technical metadata and preservation actions that I need to record.
So what is stopping me from starting to use this today? Firstly a confession – I like databases. Given the choice I would much rather work with databases than XML. This isn’t a problem in itself. Rob Sharpe from Tessella explained that a system can be externally PREMIS compliant even if PREMIS isn’t used internally. So long as the metadata recorded by an archive can be mapped to PREMIS then PREMIS XML could be exported and made available to others.
Following on from this, it should not really matter whether you prefer databases or XML so long as you have the right tools for the job. A preservation system in which much of the preservation metadata is automatically generated and where a user-friendly interface is provided for the creation and editing of preservation metadata would be the ideal. How the data is stored behind the scenes is not so important as long as you know you are storing the right bits of information.
Fortunately there are several digital preservation tools into which PREMIS is incorporated. Some of these such as Rosetta and Archivematica are already on my list of systems to look at. Armed with the knowledge I gained yesterday (and a pocket full of jelly beans) I am in a good position to push forward with my investigations.